Periodic Reporting for period 4 - TRANSEP (Flow physics and interaction of laminar-turbulent transition and flow separation studied by direct numerical simulations)
Periodo di rendicontazione: 2021-03-01 al 2021-08-31
The vision of this work is to overcome the failure of Computational Fluid Dynamics (CFD) to tackle one of the central unsolved fluid physics problems, namely predicting the sensitive flow physics associated with laminar-turbulent transition and flow separation. A recent, highly influential report by NASA clearly states that the major shortcoming of CFD is its “… inability to accurately and reliably predict turbulent flows with significant regions of separation”, most often associated with laminar-turbulent transition. The research address this shortcoming and develop and utilize computational methods that are able to predict, understand and control the sensitive interplay between laminar-turbulent transition and flow separation in boundary layers on wings and other aerodynamic bodies.
We will be able to understand enigmas such as the recent results from experiments where the laminar area of a wing grows after a smooth surface have been painted (increased roughness), or the drastic changes of laminar-turbulent transition and separation locations on unsteady wings, or the notoriously difficult interaction of multiple separation and transition regions on high-lift wing configurations. For such flows there have been little understanding of flow physics and few computational prediction capabilities. Here we will perform simulations that give completely new possibilities to visualize, understand and control the flow around such wings and aerodynamic bodies, including the possibility to compute and harness the flow sensitivities. We tackle these outstanding flow and turbulence problem using the new possibilities enabled by multi-peta scale computing.
These research questions are highly important in the design of new fuel efficient aircraft, where the high drag of the turbulent and separated flow needs to be avoided as much as possible. If laminar flow control is utilised in a modern transport aircraft a typical fuel savings of 15% is envisaged.
We will be able to understand enigmas such as the recent results from experiments where the laminar area of a wing grows after a smooth surface have been painted (increased roughness), or the drastic changes of laminar-turbulent transition and separation locations on unsteady wings, or the notoriously difficult interaction of multiple separation and transition regions on high-lift wing configurations. For such flows there have been little understanding of flow physics and few computational prediction capabilities. Here we will perform simulations that give completely new possibilities to visualize, understand and control the flow around such wings and aerodynamic bodies, including the possibility to compute and harness the flow sensitivities. We tackle these outstanding flow and turbulence problem using the new possibilities enabled by multi-peta scale computing.
These research questions are highly important in the design of new fuel efficient aircraft, where the high drag of the turbulent and separated flow needs to be avoided as much as possible. If laminar flow control is utilised in a modern transport aircraft a typical fuel savings of 15% is envisaged.
1. First turbulence simulation of flows over wing section at Re=1,000,000
We have conducted well-resolved large-eddy simulations (LESs) of the turbulent flow around a NACA4412 wing section up to a numerically high Reynolds number of 1 million (based on inflow velocity and wing chord length). We carefully designed the numerical setup, including resolution and accuracy requirements, and implemented a relaxation-term-based filter for the LES in Nek5000. The LES results were validated against fully-resolved direct numerical simulations (DNSs) of the same flow case.
2. Data-driven methods for studying turbulence
We have developed a data-driven method for predictions of turbulence quantities in spatially-developing boundary layers. In particular, we employed a system identification approach where high-fidelity numerical data are used to build single and multiple-input linear and non-linear transfer functions. The developed methodology has great potential for implementation in experiments and realistic flow control applications.
3. First high fidelity simulations of pitching airflows at moderately high Reynolds numbers
We have conducted high-fidelity simulations of unsteady wings at two different Reynolds numbers. We studied the unsteady boundary layer transition of a laminar wing undergoing forced pitch oscillations at a moderately high Reynolds number of 750,000. The study lead to the development of a low-dimensional model for the approximation of unsteady aerodynamic loads and a much improved understanding of the origin of non-linear aerodynamic response of unsteady laminar airfoils.
Another study involved the study of unsteady aerodynamics of pitching airfoils at lower Reynolds numbers (Re=100,000) which brought to light the complex dynamical behaviour of unsteady separation bubbles. The study found that unsteady separation bubbles can undergo state changes from convective to absolute instability and that such state changes cause abrupt changes in the boundary-layer characteristics of the airfoil, leading to large variations in aerodynamic loading.
4. Large scale simulations of transition under free-stream turbulence in boundary layers over flat plates and low-pressure turbine blades
We also study effects of the free-stream turbulence characteristic length scales and intensity on the transition in an incompressible flat-plate boundary layer by means of DNS. Computations are performed using the spectral element code Nek5000. Numerically-generated homogeneous isotropic turbulence upstream of the leading edge is designed to imitate the characteristics of the grid-generated turbulence in the wind tunnel experiments. Various combination of levels of the free-stream turbulence intensity and integral length scales are simulated. Turbulence statistics and integral quantities are carefully evaluated showing close agreement with the corresponding experimental data.
Further, we study both the effect of the level of free-stream turbulence and the effect of the upstream wakes on the transition in the flow over low-pressure turbine blades. In a second stage of the study, cylinders moving in front of the leading edge of the turbine are included to model the effect of the wake coming from upstream blade. The flow structures and the receptivity of the flow to continuous forcing by the free-stream turbulence and perturbations at the leading edge have been analysed using Spectral Proper Orthogonal Decomposition (SPOD).
5. Large scale simulations of roughness induced transition on swept wings.
We have investigated, through direct numerical simulations, the combined effects of an isolated roughness element together with the presence of free-stream turbulence on a swept-wing boundary layer. In absence of free-stream turbulence it is possible to observe the presence of stationary crossflow vortices behind the roughness element. For a very low free-stream turbulence level (0.03%), it is possible to observe growing disturbances behind small roughness elements that form turbulent spots while traveling downstream. If the level of free-stream turbulence is increased one order of magnitude, a clear laminar-turbulent transition takes place behind the roughness. This study has helped to understand the high sensitivity of the wake of a roughness element to the level of free-stream turbulence and explaining the experimental observations.
We have conducted well-resolved large-eddy simulations (LESs) of the turbulent flow around a NACA4412 wing section up to a numerically high Reynolds number of 1 million (based on inflow velocity and wing chord length). We carefully designed the numerical setup, including resolution and accuracy requirements, and implemented a relaxation-term-based filter for the LES in Nek5000. The LES results were validated against fully-resolved direct numerical simulations (DNSs) of the same flow case.
2. Data-driven methods for studying turbulence
We have developed a data-driven method for predictions of turbulence quantities in spatially-developing boundary layers. In particular, we employed a system identification approach where high-fidelity numerical data are used to build single and multiple-input linear and non-linear transfer functions. The developed methodology has great potential for implementation in experiments and realistic flow control applications.
3. First high fidelity simulations of pitching airflows at moderately high Reynolds numbers
We have conducted high-fidelity simulations of unsteady wings at two different Reynolds numbers. We studied the unsteady boundary layer transition of a laminar wing undergoing forced pitch oscillations at a moderately high Reynolds number of 750,000. The study lead to the development of a low-dimensional model for the approximation of unsteady aerodynamic loads and a much improved understanding of the origin of non-linear aerodynamic response of unsteady laminar airfoils.
Another study involved the study of unsteady aerodynamics of pitching airfoils at lower Reynolds numbers (Re=100,000) which brought to light the complex dynamical behaviour of unsteady separation bubbles. The study found that unsteady separation bubbles can undergo state changes from convective to absolute instability and that such state changes cause abrupt changes in the boundary-layer characteristics of the airfoil, leading to large variations in aerodynamic loading.
4. Large scale simulations of transition under free-stream turbulence in boundary layers over flat plates and low-pressure turbine blades
We also study effects of the free-stream turbulence characteristic length scales and intensity on the transition in an incompressible flat-plate boundary layer by means of DNS. Computations are performed using the spectral element code Nek5000. Numerically-generated homogeneous isotropic turbulence upstream of the leading edge is designed to imitate the characteristics of the grid-generated turbulence in the wind tunnel experiments. Various combination of levels of the free-stream turbulence intensity and integral length scales are simulated. Turbulence statistics and integral quantities are carefully evaluated showing close agreement with the corresponding experimental data.
Further, we study both the effect of the level of free-stream turbulence and the effect of the upstream wakes on the transition in the flow over low-pressure turbine blades. In a second stage of the study, cylinders moving in front of the leading edge of the turbine are included to model the effect of the wake coming from upstream blade. The flow structures and the receptivity of the flow to continuous forcing by the free-stream turbulence and perturbations at the leading edge have been analysed using Spectral Proper Orthogonal Decomposition (SPOD).
5. Large scale simulations of roughness induced transition on swept wings.
We have investigated, through direct numerical simulations, the combined effects of an isolated roughness element together with the presence of free-stream turbulence on a swept-wing boundary layer. In absence of free-stream turbulence it is possible to observe the presence of stationary crossflow vortices behind the roughness element. For a very low free-stream turbulence level (0.03%), it is possible to observe growing disturbances behind small roughness elements that form turbulent spots while traveling downstream. If the level of free-stream turbulence is increased one order of magnitude, a clear laminar-turbulent transition takes place behind the roughness. This study has helped to understand the high sensitivity of the wake of a roughness element to the level of free-stream turbulence and explaining the experimental observations.
The progress beyond the state of the art is contained in three main results of the project:
Incorporation of the possibility to perform unsteady aerodynamics in the Nek5000 open access software and the use of this software to understand the fluid physics of the transition and separation processes associated with the pitching airfoil study.
Understading of the roughness induced instability in the swept wing RECEPT experiment as a localised wave packet growth, giving an earlier instability compared to fully turbulent transition at the roughness location.
The first fully wall resolved accurate LES simulations of a 1.000.000 Reynolds number turbulent wing ever performed.
Development of a linear framework for fluid-structure interaction. For the first time, we have determined the bifurcation to an unsteady fluid interaction through a global instability calculation for a realistic wing.
Detailed investigation of competing receptivity mechanisms in flow over a low-pressure turbine blade through SPOD analysis of data from our direct numerical simulations.
Incorporation of the possibility to perform unsteady aerodynamics in the Nek5000 open access software and the use of this software to understand the fluid physics of the transition and separation processes associated with the pitching airfoil study.
Understading of the roughness induced instability in the swept wing RECEPT experiment as a localised wave packet growth, giving an earlier instability compared to fully turbulent transition at the roughness location.
The first fully wall resolved accurate LES simulations of a 1.000.000 Reynolds number turbulent wing ever performed.
Development of a linear framework for fluid-structure interaction. For the first time, we have determined the bifurcation to an unsteady fluid interaction through a global instability calculation for a realistic wing.
Detailed investigation of competing receptivity mechanisms in flow over a low-pressure turbine blade through SPOD analysis of data from our direct numerical simulations.